JPH0722109B2 - Method for determining light exposure of photosensitive rack layer - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is for obtaining a lacquer pattern on a substrate which needs to be reproduced exactly after exposing through a mask and then developing the details of the photosensitive lacquer having a line width in the range of 0.5-2 μm. In particular, it relates to a method for determining the exposure dose of a photosensitive lacquer layer present on said substrate.

In order to obtain a lacquer pattern with a very precise line width in the case of lines in the range 0.5-2 μm, an accurate exposure dose, usually an exposure at an exact exposure time, is required. If the exposure amount is not proper, the line width will be displaced. In recent semiconductor technology, it is extremely important to process extremely small dimensions with extremely high accuracy.

Various difficulties are encountered in the micron range described above when measuring the line width obtained after exposure for different periods of time to determine the proper exposure to obtain the desired line width. An optical microscope is used in the conventional measuring method.
However, in this case the reliability is relatively low. The line width in the above-mentioned range is at the limit of the possibility of the optical microscope, and if the thickness of the lacquer pattern is significantly increased, the thickness of the lacquer pattern may exceed the depth of field of the microscope. Another possible method of determining the appropriate exposure is to measure the line width obtained by scanning electron microscopy. However, for this purpose, it is necessary to prepare such expensive equipment, and it is necessary to satisfy the necessary conditions when performing such measurement using a scanning electron microscope.

The purpose of the invention is to determine the standard exposure in a simple manner and at the same time to influence the exposure on the processing conditions such as the thickness of the lacquer layer, the photosensitivity of the lacquer, the temperature of the oven during the baking of the lacquer. The purpose is to provide a method for automatically compensating for.

In the present invention, in the method described at the outset, a photosensitive lacquer layer present on a substrate is exposed through a raster pattern of alternating transparent and opaque linear regions having a desired width, and then the raster pattern is lined. After moving laterally in a direction perpendicular to the length of the line for a distance equal to the width, the exposure is again carried out with the same exposure, the process is repeated in different areas with different exposures and then after development the photosensitive lacquer layer is applied. The above-mentioned object is achieved by a method for determining the exposure amount of a photosensitive lacquer layer, which comprises determining the exposure amount required for exactly complete dissolution and determining the standard exposure amount.

The invention is based on the finding that in the transition region of the raster from transparent to opaque, the intensity of the light incident on the lacquer layer is smaller due to scattering and spots in the transparent part. The depth of the light penetrating into the lacquer layer in such a transient region depends on the exposure time. As in the case of the present invention, relatively short exposures occur when a raster of the desired minimum line width is exposed, and then the raster is moved over the line width and then re-exposed at different exposure doses in different areas, for example at exposure times. The lacquer beam remains after development of the lacquer layer at the paddle exposure time. Such a lacquer beam is clearly visible in an ordinary microscope due to scattering and refraction. If the exposure time is long, the lacquer beam is suddenly absent. The exposure dose associated with the development in which the lacquer beam disappears rapidly indicates the standard exposure dose.

If the experimental exposure and development are performed under the same processing conditions as those used for processing semiconductor wafers for manufacturing purposes, the processing conditions do not shift. The standard exposure dose only needs to be determined once for a given lacquer and then dictates the exposure to the lacquer for semiconductor wafers processed for manufacturing.

During the processing of semiconductor wafers, exposure is performed with a standard exposure dose, that is, an exposure dose at which the lacquer beam suddenly disappears, but the line width obtained during this process does not match the desired line width. However, when the exposure time associated with the desired line width is determined, the difference between the exposure time for standard exposure and the exposure time for the measured proper line width and exposure is , Is constant over the entire range of desired line width. In the example of the present invention utilizing this fact, for a given type of photosensitive lacquer, the exposure required to obtain the desired line width was determined and then compared with the standard exposure obtained under the same processing conditions. The difference is calculated, and thereafter, for each lot of the substrate to be processed, the exposure amount is selected so as to be equal to the total of the standard exposure amounts under the processing conditions to be selected, and the value of the difference is calculated.

The invention will now be described by way of example with reference to the drawings.

FIG. 1 shows a raster consisting of a pattern of transparent line-shaped areas 1 and opaque line-shaped areas 2, which raster is provided on a support 3, for example of glass or a transparent synthetic material. Regions 1 and 2 have the same width.
Very small dimensions precisely formed on the photo tracker, i.e. 0.
To determine the exposure dose for a substrate coated with a photolacquer layer in order to obtain a line width of 5 to 2 μm, use a number of line widths with different regions 1 and 2 and a large number of lines for each line width. An exposure experiment can be conducted by selecting exposures having different intensities.

In the raster shown in FIG. 1, the width of the transparent area 1 and the width of the opaque area 2 are each 1 μm. Rasters are provided on the support 3 by means of, for example, a step-and-repeat camera in the areas on the semiconductor wafer where it is customary to place masks for the exposure of the tracker. The raster is then exposed, for example with a 120 millisecond flash, and the raster pattern is projected onto a semiconductor wafer phototracker, for example in a ratio of 1: 1.

FIG. 2 shows a portion of a substrate 4 coated with a phototracker, for example the positive phototracker HPR204 (trade name) from Hunt, exposing this phototracker through a raster pattern. The exposed area of the photo tracker is shown at 5, and the unexposed area is shown at 6. Then, the moving table on which the substrate 4 such as a semiconductor wafer is mounted is moved at a right angle to the length direction of the regions 5 and 6 over a distance equal to 1/2 of the pitch of these two regions, in this example, 1 μm. The substrate 4 is then exposed again to the phototracker with a 120 millisecond flash. A second exposure is applied to the exposed portion 5 of the phototracker obtained by the first exposure through the opaque areas 2 of the raster. FIG. 3 shows this step. The second exposure area is indicated by reference numeral 7. Further, similar exposure is performed by changing the exposure time in a large number of areas, for example, 130 milliseconds, 140 milliseconds, and the like up to 180 milliseconds.

Development of the exposed and baked positive photo tracker dissolves the exposed areas. Two exposures completely expose the experimental area, but when the exposure time is relatively short, not the entire amount of phototracker is dissolved, but a transient between the transparent and opaque portions of the raster beam 8 during projection. Area trackers remain (see Figure 4). This is because the light cannot scatter and refract into the bottom of the phototracker layer with sufficient strength. Such a raster beam 8 can be clearly observed with an ordinary microscope. It was found that the raster beam suddenly disappeared when the exposure time was long. The relevant exposure time is the standard exposure for the photosensitive layer used in the processing conditions used. This standard exposure is easily determined and is the basis for determining the exposure to obtain a line-shaped area in the phototracker with a very precise value, in the example above of 1 μm.

For example, during the manufacture of a semiconductor device, when the phototracker layer on the wafer is exposed at the above-mentioned exposure amount through a mask having a line-shaped region occupying a width of, for example, 1 μm when projected onto the phototracker, The areas formed after development do not exhibit the desired 1 μm width. If the line width is extremely accurately determined by an experiment to obtain an exposure amount having a desired value (for example, by a scanning electron microscope), 0.5 to 2 μm
It was found that the difference between the required exposure amount and the standard exposure amount was a constant value in all widths of the range. For a given type of lacquer, this difference only needs to be determined once.
For each desired line width, the sum of the standard exposure amount to be obtained and the difference value once obtained is the required exposure amount. In the case of the above-mentioned Photocurracer HPR204, the value of the difference was experimentally found to be 60 ms. If the standard exposure for the 1 .mu.m region is determined to be 170 ms, then the exposure time required to accurately obtain a 1 .mu.m wide region in the lacquer layer is 170 + 60 = 230 ms.

[Brief description of drawings]

1 is a front view of a raster having alternating transparent and opaque regions, FIG. 2 is a schematic diagram showing a portion of a substrate having a photosensitive lacquer layer exposed through the raster of FIG. 1, FIG. Is a schematic diagram of the substrate of FIG. 2 after a second exposure, and FIG. 4 is a schematic diagram of the substrate of FIG. 3 still having a lacquer beam after development. 1 ... Transparent line-shaped area 2 ... Opaque line-shaped area 3 ... Support 4 ... Substrate 5 ... Phototracker exposed area (exposed area) 6 ... Phototracker non-exposed area 7 ... Second exposure area 8 ... Raster beam

Claims (2)

[Claims] 1. Exposure through a mask, then 0.5-2 .mu.
In order to obtain a lacquer pattern on the substrate which needs to be reproduced exactly after developing the details of the photosensitive lacquer having a line width in the range of m, the exposure of the photosensitive lacquer layer present on said substrate is determined. Exposing a photosensitive lacquer layer present on the substrate through a raster pattern of alternating transparent and opaque linear regions having a desired width, and then applying the raster pattern to the line length over a distance equal to the line width. After traversing in a direction perpendicular to the direction, the exposure is again carried out with the same exposure, this process is repeated in different areas with different exposures, and then it is necessary to just completely dissolve the photosensitive lacquer layer after development. A method for determining the exposure amount of a photosensitive lacquer layer, which comprises determining the exposure amount and determining the standard exposure amount. 2. For a given type of photosensitive lacquer, the exposure dose required to obtain the desired line width is determined, and then the difference from the standard exposure dose obtained under the same processing conditions is determined. 2. The amount of exposure is selected for each lot of substrates to be processed so as to be equal to the sum of standard exposures under the processing conditions to be selected, and the value of the difference is obtained. The method described.